I’d like to know if a flac audio file showing 21000Hz as shown in the attached image is regarded as a true lossless file? I ask as to me there appears to be a dramatic cut-off and not a slow cut-off.
The term “lossless” in relation to audio formats, means that the round trip from raw data to file back to raw data can return the exact same data. Of course this will only be true if no other changes are made during the round trip. In this sense, FLAC is a lossless format.
It is important to note that just as you can have a good quality recording in a lossless format, you can also have a bad quality recording in a lossless format. The difference between bad audio in a lossy format and bad audio in a lossless format, is that the lossless format can be decoded to the same bad audio data that you encoded, whereas a lossy format will give you slightly different bad audio.
Thanks for your reply.
I really wanted your advice on that particular frequency which shows in the image that I provided.
Note that microphones tend to have a steep dropoff at that range as well:
Ok,let me explain, this image is from an audio file with a flac extension , all I really wish to know is does the plot spectrum analysis look like it is a lossless audio file?
I have just done a check using a Lossless Audio Checker version 2.0.7 and it reports the result as Clean meaning it is a true uncompressed lossless audio file.
That does not tell us anything about the quality of the audio. It just tells us that the frequencies roll off steeply from a little over 20 kHz. There could be any number of reasons why it does that. It is not at all unusual for professional recordings to roll off around 20 kHz.
Thanks for the info.
I just wanted to ensure that it hadn’t been upsampled, lossless checker says it’s clean so hasn’t been upsampled.
The spectrum is a CLUE, and it’s a pretty good clue with MP3s but you can get false-positives and false-negatives and it doesn’t prove anything.
Lossless Audio Checker is telling you if it’s PROBABLY from lossy or not.
MP3 generally cuts-off between 16 & 17kHz, but it can be tweaked. Lossy compression has to throw-away SOMETHING and if you force it keep the higher frequencies the spectrum will look better but it might throw away something more important and sound might be worse!
It’s easier to make a nice spectrum than a good sound!
With a good quality (high-bitrate) MP3, if you hear a compression artifact, or a difference from the original, it’s usually NOT the loss of high frequencies.
I think other compressed formats keep more of the highs.
It’s EASY to make a “prettier” spectrum and fool lossless audio checker. After converting to WAV or FLAC you can run an “exciter” or enhancer to add high frequencies. But you can’t fix “pre-echo” or other compression artifacts that can’t be as easily measured.
This topic was automatically closed after 30 days. New replies are no longer allowed.